Trace picture piles on layers

Trace picture piles on layers

R=enne

Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=200154

[nduca, 2013-05-14 15:12:40]

ptal

I'm not sure I've got this right. When a layer is completely clobbered, its
picture actually looks right. But when there are small updates, I'm getting
strangely stretched pictures... they look like they're in the right place, but
the contents are funky. That might just be a bug in my UI side, but figured i'd
mention it.

[enne (OOO), 2013-05-14 15:55:18]

lgtm

I don't see anything wrong with what you're doing here.

https://codereview.chromium.org/14772032/diff/1/cc/base/math_util.cc
File cc/base/math_util.cc (right):

https://codereview.chromium.org/14772032/diff/1/cc/base/math_util.cc#newcode531
cc/base/math_util.cc:531: res->AppendDouble(r.x());
AppendInteger?

https://codereview.chromium.org/14772032/diff/1/cc/resources/picture_pile_bas...
File cc/resources/picture_pile_base.cc (right):

https://codereview.chromium.org/14772032/diff/1/cc/resources/picture_pile_bas...
cc/resources/picture_pile_base.cc:180: scoped_ptr<base::ListValue> rects(new
base::ListValue());
rects => pictures?

https://codereview.chromium.org/14772032/diff/1/cc/resources/picture_pile_bas...
cc/resources/picture_pile_base.cc:181: gfx::Rect
layer_rect(tiling_.total_size()); /* enne, insane or not? */
This is right.  You could also simplify this a little by just iterating through
the picture_list_map_ if you wanted.

[vmpstr2, 2013-05-14 18:16:09]

https://codereview.chromium.org/14772032/diff/1/cc/debug/traced_picture.cc
File cc/debug/traced_picture.cc (right):

https://codereview.chromium.org/14772032/diff/1/cc/debug/traced_picture.cc#ne...
cc/debug/traced_picture.cc:35: picture_->LayerRect().x(),
picture_->LayerRect().y(),
picture's layer rect is encoded as a part of AsBase64String. Is the reason for
putting it here separately just so we can tell what the layer is without
decoding the picture? (I'm just trying to catch up on understanding)

[enne (OOO), 2013-05-14 20:47:52]

On 2013/05/14 18:16:09, vmpstr2 wrote:
> https://codereview.chromium.org/14772032/diff/1/cc/debug/traced_picture.cc
> File cc/debug/traced_picture.cc (right):
> 
>
https://codereview.chromium.org/14772032/diff/1/cc/debug/traced_picture.cc#ne...
> cc/debug/traced_picture.cc:35: picture_->LayerRect().x(),
> picture_->LayerRect().y(),
> picture's layer rect is encoded as a part of AsBase64String. Is the reason for
> putting it here separately just so we can tell what the layer is without
> decoding the picture? (I'm just trying to catch up on understanding)

Yeah, that's exactly it.

[nduca, 2013-05-14 21:21:31]

The js needs to know the picture params. We could have done this by adding more
methods to skia_benchmarking_extension, but that means if you dont have the
extension installed, you can't even know where the picture is located on the
layer, which is kinda whommp.

[commit-bot: I haz the power, 2013-05-14 21:40:40]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/nduca@chromium.org/14772032/9001

[commit-bot: I haz the power, 2013-05-15 04:33:34]

Change committed as 200154